WO2007009958A1 - Terpene synthases - Google Patents

Abstract

The present invention is directed to new terpene synthases, the nucleic acids encoding same as well as to inhibitors of said terpene synthases. The invention is further directed to a method of producing a transgenic plants and of producing secondary plant metabolites. The invention further discloses essential oils produced by said method as well as pharmaceutical and food products containing same.

Description

Terpene Synthases

The present invention is directed to new terpene synthases, the nucleic acids encoding same as well as to inhibitors of said terpene synthases. The invention is further directed to a method of producing transgenic plants and of producing secondary plant metabolites. The invention further discloses essential oils produced by said method as well as pharmaceutical and food products containing same.

Plant secondary metabolites are those compounds produced by plants that are not directly essential for the basic photosynthetic or respiratory metabolism. Terpenes or terpenoids are the largest and most widespread class of secondary metabolites. Over 22,000 different structures have been described in 1991 by Connolly and Hill. In 1998, this number rose to over 30,000 different structures (Buckingham 1998) which is an addition of over 1,300 newly described and characterized terpenes per year. Despite the vast amount of terpene structures, they can be traced back to a relatively simple biosynthesis from 5-carbon units. Terpenes are classified based on the number of C₅ isoprenoid units in their structures (Table 1). Monoterpenes consist of two, sesquiterpenes of three and polyterpenes from nine up to 30,000 connected isopentenoid units. Terpenoids have not only an enormous structural variability but also a great functional diversity. They play important roles in almost all basic plant processes, including growth, development, reproduction and defense (Gershenzon and Kreis, 1999).

Table 1 : Classification of terpenoids.

Terpenoids follow us all day long as components in perfumes and showering lotions, as the smell of a nice bouquet of flowers or from herbs and spices in our food. Therefore, much effort is being devoted to finding new terpenoids or new varieties of better smelling or tasting herbs. One of the most common herbs include sage, oregano, thyme and basil, which all have become commercially important herbs in the USA (Olivier, 1997).

The biosynthesis of terpenes takes place in the head cells of the glandular trichomes (McCaskille/ al., 1992; Gershenzon et al., 1989). Two pathways of isoprenoid production are present in plants; the mevalonic acid (MVA) pathway for sesquiterpenes, and the methyl-erythritol-phosphate (MEP) pathway for monoterpenes and diterpenes. Both pathways form isopentenyl pyrophosphate (IPP) which is the precursor for all terpenes. The MVA pathway is localized in the cytosol and the MEP pathway in the plastids (Lichtenthaler et al., 1997).

The MVA pathway forms IPP from three molecules of Acetyl-CoA and was first described by Qureshi and Porter (1981) whereas the MEP pathway was first reported as a non-mevalonate dependent pathway by Flesh and Rohmer (1988). Both pathways have been reviewed by Lichtenthaler et al. (1997) (Figure 1). It is thought that the biosynthesis of monoterpenes is localized in plastids and utilizes the IPP product of the MEP pathway.

The IPP can be converted by an isomerase to dimethylallyl pyrophosphate (DMAPP). Prenyltransferases use one molecule DMAPP and IPP to form geranyl pyrophosphate (GPP), the precursor of all monoterpenes (Gershenzon and Croteau, 1993) (Fig. 2).

The conversion of GPP to a large variety of terpene hydrocarbon skeletons is catalyzed by the enzyme class of terpene synthases which employ a carbocationic reaction mechanism and often introduce cyclizations (Wise and Croteau, 1999). The geranyl pyrophosphate is ionized into a geranyl cation which is transformed into the enzyme-bound linalyl pyrophosphate (Figure 3). This linalyl pyrophosphate is transformed from the transoid to the cisoid form via rotation of the C₂-C₃ single bond. The cyclisation is reached upon the juxtaposition of one bond of the C₆=C₇ double bond. The resulting α-terpinyl cation is the precursor for all monocyclic and bicyclic monoterpenes. Acyclic monoterpenes like myrcene are formed from earlier intermediates on the way from GPP to the α-terpinyl cation.

Since essential oils are of growing interest in various fields of science and commerce, e.g. in the pharmaceutical and food industry, there is an ongoing need for further enhanced methods of producing those oils in high amounts and quality. However, the production of essential oils and in particular the active ingredients (as the terpenes mentioned above) contained therein by culturing plants and obtaining oils by steam or hydrodistillation is often time consuming and difficult. Therefore, it is an object of the present invention to provide improved means and methods for improving biosynthesis of terpenes. It is in particluar an object of the invention to provide improved means and methods for obtaining higher amounts and/or a higher degree of quality of terpenes in plants, in particular of mono- and sesquiterpenes. It is a further object of this invention to provide a transgenic plant which is capable of producing those terpenes in a high amount.

These objects are achieved by the subject-matter of the independent claims. Preferred embodiments are set forth in the dependent claims.

The original aim of the inventors was to identify and biochemically characterize enzymes that are responsible for the formation of terpenes in two Origanum vulgare clonal lines with different terpene composition in their essential oils. A key step of terpene biosynthesis is the formation of the large variety of mono- and sesquiterpene skeletons from the precursors geranyl pyrophosphate and farnesyl pyrophosphate, respectively. These reactions are catalyzed by the enzyme class of terpene synthases. To study terpene biosynthesis and its diversity in two Origanum lines, f02-04 and d06-01 , terpene synthases were isolated and characterized.

As a result, new terpene synthases were identified, which are termed TPS herein. The enzymes TPSl and TPS2, as an example, produced 57 % of the terpenes in the characteristic relative proportions found in the originating plant line. Thus, TPSl and in particular TPS2 may serve as a valuable tool for improving the biosynthesis of terpenes (or essential oils) by, for example, generation of transgenic plants containing and expressing the relevent nucleic acid sequences.

In particular, the present invention is directed to the following aspects and embodiments:

According to a first aspect, a terpene synthase is provided, which is encoded by the nucleic acid of SEQ ID NO: 1 or 2 or variants thereof, which variants are each defined as having one or more substitutions, insertions, and/or deletions as compared to the nucleic acid of SEQ ID NO: 1 or 2, provided that:

a) these variants hybridize under moderately stringent conditions to a nucleic acid, which comprises the sequence of SEQ ID NO: 1 or 2, and further provided that these variants code for a protein having terpene synthase activity; or b) these variants have nucleic acid changes which are due to the degeneration of the genetic code, which code for the same or functional equivalent amino acid as the nucleic acid of SEQ ID NO: 1 or 2.

Furthermore, in a second aspect, an isolated nucleic acid is provided, which comprises the nucleic acid of SEQ ID NO: 1 or 2 or variants thereof, wherein the variants are each defined as having one or more substitutions, insertions, and/or deletions as compared to the nucleic acid of SEQ ID NO: 1 or 2, provided that:

a) these variants hybridize under moderately stringent conditions to a nucleic acid, which comprises the sequence of SEQ ID NO: 1 or 2, and further provided that these variants code for a protein having terpene synthase activity; or

b) said variants have nucleic acid changes which are due to the degeneration of the genetic code, which code for the same or functional equivalent amino acids as the nucleic acid of SEQ ID NO: 1 or 2.

The nucleic acid variants according to the invention also comprise nucleic acid fragments which contain more than 10, preferably more than 15, more than 20, more than 25 or more than 30 and up to 50 nucleotides. The term oligonucleotide includes fragments containing 10 to 50 nucleotides and parts thereof. These sequences can be in any order as long as at least 10 successive nucleotides are according to the invention. These oligonucleotides can be preferably used as primer, for example for RT-PCR or as a probe for in situ hybridization.

According to the state of the art an expert can test which derivatives and possible variations derived from these revealed nucleic acid sequences according to the invention are, are partially or are not appropriate for specific applications like hybridization and PCR assays. The nucleic acid and oligonucleotides of the inventions can also be part of longer DNA or RNA sequences, e.g. flanked by restriction enzyme sites.

Amplification and detection methods are according to the state of the art. The methods are described in detail in protocol books which are known to the expert. Such books are for example Sambrook et al., 1989, Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Press, and all subsequent editions. PCR-methods are described for example in Newton, PCR, BIOS Scientific Publishers Limited, 1994 and all subsequent editions. As defined above, "variants" are according to the invention especially such nucleic acids, which contain one or more substitutions, insertions and or deletions when compared to the nucleic acids of SEQ ID NO: 1 and 2. These lack preferably one, but also 2, 3, 4, or more nucleotides 5' or 3' or within the nucleic acid sequence, or these nucleotides are replaced by others. The nucleic acid sequences of the present invention comprise also such nucleic acids which contain sequences in essence equivalent to the nucleic acids described in SEQ ID NO: 1 and 2. According to the invention nucleic acids can show for example at least about 80%, more typically at least about 90% or 95% sequence identity to the nucleic acids described in SEQ ID NO. 1 and 2.

The term "nucleic acid sequence" means a heteropolymer of nucleotides or the sequence of these nucleotides. The term "nucleic acid", as herein used, comprises RNA as well as DNA including cDNA, genomic DNA and synthetic (e.g. chemically synthesized) and to other polymers linked bases such as PNA (peptide nucleic acids).

The invention comprises - as mentioned above - also such variants which hybridize to the nucleic acids according to the invention at moderate stringent conditions. The invention, however, also comprises such variants which hybridize to the nucleic acids according to the invention at conditions of stringency or high stringency.

Stringent hybridization and wash conditions are in general the reaction conditions for the formation of duplexes between oligonucleotides and the desired target molecules (perfect hybrids) or that only the desired target can be detected. Stringent washing conditions mean 0.2 x SSC (0.03 M NaCl, 0.003 M sodium citrate, pH 7)/0.1% SDS at 65°C. For shorter fragments, e.g. oligonucleotides up to 30 nucleotides, the hybridization temperature is below 65°C, for example at 50⁰C, preferably above 55⁰C, but below 65⁰C. Stringent hybridization temperatures are dependent on the size or length, respectively of the nucleic acid and their nucleic acid composition and will be experimentally determined by the skilled artisan. Moderate stringent hybridization temperatures are for example 42⁰C und washing conditions with 0.2 x SSC/0.1% SDS at 42°C.

The respective temperature conditions can vary dependent on the chosen experimental conditions and to be tested nucleic acid probe, and have to be adapted appropriately. The detection of the hybridization product can be done for example using X-Ray in the case of radioactive labeled probes or by fluorimetry in the case of fluorescent labeled probes.

The expert can according to the state of the art adapt the chosen procedure, to reach actually moderate stringent conditions and to enable a specific detection method. Appropriate stringent conditions can be determined for example on the basis of reference hybridization. An appropriate nucleic acid or oligonucleotide concentration needs to be used. The hybridization has to occur at an appropriate temperature (the higher the temperature the lower the binding).

Fragments of the nucleic acids according to the invention can be used for example as oligonucleotide primer in detection systems and amplification methods of the tps gene and transcript. The expert can apply these oligonucleotides in state of the art methods. DNA or RNA can be analyzed for the presence of one of the described genes or transcripts applying the appropriate oligonucleotide primers to the to be analyzed probe. The detection of the RNA or DNA of the probe can be achieved for example by PCR methods, which reveal the presence of the specific DNA and/or RNA sequences. All hereinabove described oligonucleotides can also be used as primers, also as primers for reverse transcription of RNA.

The PCR method has the advantage that very small amounts of DNA are detectable. Dependent on the to be analyzed material and the equipment used the temperature conditions and number of cycles of the PCR have to be adjusted. The optimal conditions can be experimentally determined according to standard procedures.

In an embodiment, the isolated nucleic acid as defined above is operably linked to one or more regulatory sequences.

The present invention comprises further transcriptional products of the hereinabove described nucleic acids and nucleic acids, which selectively hybridize under moderate stringent conditions to one of these transcriptional products. Preferably this comprises an antisense DNA or RNA in form of a DNA or RNA probe which can hybridize to a transcription product, e.g. mRNA, and can be used in detection systems.

Furthermore, siRNA is a preferred transcriptional product for SEQ ID NO: 1 or 2. RNA interference in general is defined as a gene silencing phenomenon whereby double-stranded RNAs trigger the specific degradation of a homologous mRNA. The specific dsRNAs are processed into small interfering RNA (siRNA) which serves as a guide for cleavage of the homologous mRNA in the RNA- induced silencing complex (RISC). This technique is useful as an application for specific suppression of an individual gene. Thus, by means of this technique it might be possible also to suppress the production of TPS 2 (or also other TPS) in transgenic plants, for example, when an altered composition of an essential oil is desired. According to a further aspect, primer for the amplification of the nucleic acids as defined herein are provided (see also above).

In a further embodiment, the present invention includes a vector (construct) comprising a nucleic acid according to the invention. This vector is preferably an expression vector which contains a nucleic acid according to the invention and one or more regulatory nucleic acid sequences.

In a preferred embodiment, the vector is a plant specific plasmid, preferably the tumor inducing (Ti) plasmid.

Furthermore, the invention provides in a further aspect a host cell, which has been transformed with the above vectors. The host cell preferably is a procaryotic cell, more preferably E.coli, B. subtilis or an agrobacterium. Here, Agrobacterium tumefaciens is in particular preferred.

In a further aspect, the invention provides a terpene synthase, comprising the amino acid sequence of SEQ ID NO: 3 or 4 or a variant of said amino acid sequences, which variant comprises one or more substitutions, insertions, and/or deletions as compared to the sequence of SEQ ID NO: 3 or 4, and wherein the biological activity of the variant is substantially equal to the activity of the terpene synthase comprising the unmodified amino acid sequence of SEQ ID NO: 3 or 4.

In particular variants of the protein, for example deletions, insertions and/or substitutions in the sequence, which cause for so-called "silent" changes, are considered to be part of the invention.

For example, such changes in the nucleic acid sequence are considered to cause a substitution with an equivalent amino acid. Preferably are such amino acid substitutions the result of substitutions which substitute one amino acid with a similar amino acid with similar structural and/or chemical properties, i.e. conservative amino acid substitutions.

Amino acid substitutions can be performed on the basis of similarity in polarity, charges, solubility, hydrophobic, hydrophilic, and/or amphipathic (amphiphil) nature of the involved residues. Examples for hydrophobic amino acids are alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan and methionine. Polar, neutral amino acids include glycine, serine, threonine, cysteine, thyrosine, asparagine and glutamine. Positively (basic) charged amino acids include arginine, lysine and histidine. And negatively charged amino acids include aspartic acid and glutamic acid. "Insertions" or "deletions" usually range from one to five amino acids. The allowed degree of variation can be experimentally determined via methodically applied insertions, deletions or substitutions of amino acids in a polypeptide molecule using recombinant DNA methods. The resulting variants can be tested for their biological activity.

Nucleotide changes, which affect the N-terminal and C-terminal part of the protein, often do not change the protein activity, because these parts are often not involved in the biological activity. It can be desired to eliminate one or more of the cysteine of the sequence, since cysteines can cause the unwanted formation of multimers when the protein is produced recombinant. Multimers may complicate purification procedures. Each of the suggested modifications is in range of the current state of the art, and under the retention of the biological activity of the encoded products.

In a further aspect, the present invention discloses an inhibitor of the terpene synthase as defined above, wherein said inhibitor is selected from the group consisting of a molecule that reduces the level of mRNA encoding said terpene synthase, a molecule that reduces the level of said terpene synthase protein, and a molecule that reduces the biological activity of said terpene synthase.

This inhibitor preferably is selected from the group consisting of an antisense nucleic acid, a ribozyme, double stranded RNA, preferably siRNA (see also above), an antibody, a peptide and a peptidomimetic.

In a further aspect, the present invention provides a method of producing a transgenic plant comprising the steps of: a) providing a nucleic acid or a vector as defined above; b) transforming plant cells or tissues with said nucleic acid or vector; and c) generating whole plants from said transformed plant cells or tissues.

In a preferred embodiment, the vector used in a) is a Ti plasmid, wherein the tumor inducing sequences were replaced by the nucleic acid of the invention (SEQ ID NO: 1 or 2). More preferably, the plant cells or tissues are transformed by infection withA tumefaciens containing the Ti plasmid.

As an alternative, the transformation in step b) may be performed by means of the gene cannon method.

In the following, the methods for producing transgenic plants are shortly outlined: A. tumefaciens is a soil bacterium which contains an extra circular chromosome called the tumor- inducing (Ti) plasmid. That DNA contains genes that are responsible for the crown gall disease of plants. It is possible to remove the genes that cause the tumors and replace them with selected genes, making the Ti plasmid a vector to transfer new genes into the plant. This method is a standard procedure well known in the art.

In vivo, infection requires wounding of the plant tissue. A. tumefaciens attaches to plant cell walls activated by compounds from the wounded cells (compounds that activate the bacteria are also produced by the wounded cells). A part of the Ti plasmid (the T-region) is then transferred into the chromosomes of the host plant where it becomes integrated (T-DNA). Several gene loci on the bacterial chromosome and a set of virulence (vir) genes located on the Ti plasmid code for functions involved in plant cell recognition and attachment as well as for the excision, transfer and integration of T-DNA into the target genome.

There are other methods for transforming plants available, e.g. the gene cannon method. Therein, minute metal beads coated with DNA are 'shot' directly into plant cells. The plant cells repair the wounds quickly and in some cells DNA is incorporated into the plant chromosomes.

In an embodiment, the plant cell or tissue is derived from terpene producing plants. Preferably, the plant is selected from sage {Salvia officinalis), oregano {Origanum vulgare), thyme {Thymus vulgaris) or basil {Ocimum basilicum).

In a further aspect, the invention pertains to a transgenic plant obtainable by the method as outlined above.

Moreover, the invention comprises a method of producing one or more secondary plant metabolites comprising the steps of: a) providing a transgenic plant as defined above, b) culturing the plant under suitable conditions; and c) recovering said one or more secondary plant metabolites from said plant.

It is noted that the methods of culturing the plant and recovering the metabolites are conventional and any of the prior art methods may be used for this purpose. For example, essential oil is distilled from plant material using water, steam or both. The process involves placing plant material over a grate and forcing steam through the grate. The plant material may also be placed in boiling water. In either case, the heat from the steam and/or water causes the plant material to break apart and release the essential oils. This vapor is condensed through cooling tubes into liquid form.

As mentioned above, the secondary metabolite is preferably selected from the group of terpenoids. Those terpenoids are comprising hemi-terpenes, mono-terpenes, sesquiterpenes, di-terpenes, tri- terpenes, tetra-terpenes and poly-terpenes (see also the information given in the introduction of the description).

In a further aspect, the invention is directed to an essential oil, which is obtainable by the method as disclosed herein. Furthermore, a pharmaceutical composition is disclosed comprising a therapeutically effective dose of said essential oil in combination with a pharmaceutically acceptable carrier or diluent. Also, a food or cosmetic product comprising the essential oil is comprised by the present invention.

Most essential oils have medicinal properties and are widely used today. For example, most essential oils have antiseptic properties, though some are stronger than others. Some essential oils are used as repellants or perfumes due to their aromatic character. Others are used (for the same reason) in aromatherapy as a form of the alternative medicine.

The present invention will be further described with reference to the following figures and examples; however, it is to be understood that the present invention is not limited to such figures and examples.

Figure 1 : A diagram of the MVA and the MEP pathway and their supposed compartimen-tation in the plant cell.

Figure 2: The proposed mechanism for the formation of GPP in plastids.

Figure 3: Proposed mechanism for the formation of the α-terpinyl cation, the precursor of monocyclic and bicyclic monoterpenes.

Figure 4: The density of glandular trichomes per cm² on young (< 1 week), middle aged (1 - 3 weeks) and old (> 4 weeks) leaves of the Origanum lines d06-01 and fU2-04.

Figure 5: GC-MS chromatograms of pentane extracts from dried young leaves from the lines dO6-

01 and f02-04.

Figure 6: Comparison of the terpene composition of the both Origanum lines.

Figure 7: A glandular trichome isolated from young Origanum leaves

Figure 8: Protein sequence of TPSl from Origanum d06-01.

Figure 9: Protein sequence alignment of both TPS2 genes performed with CLUSTALX (1.83). Figure 10: GC-MS chromatogram of terpene products of the truncated TPSl enzyme

Figure 11 : GC-MS chromatogram of terpene products of TPS2 enzyme.

Figure 12: Time course of product formation for TPS2.

Figure 13: Enzyme kinetics with two different protein concentrations for increasing concentrations of GPP substrate.

Figure 14: Enzyme kinetics in dependence to Mn²⁺ ions.

Figure 15: Enzyme kinetics in dependence OfMg²⁺ ions

Figure 16: Enzyme activity at different pH values.

Figure 17: Enzyme activity at different temperatures.

Figure 18: Comparison of the terpene blend of leaf pentane extracts from the plant line with the products of heterologously expressed γ-terpinene synthase (TPS2).

Figure 19: Transcript levels oitps2 in two plant lines.

Figure 20: Unrooted dendrogram of the amino acid sequences of 48 monoterpene synthases.

Figure 21 : Proposed reaction mechanism of TPS2.

Figure 22 GC-MS chromatogram of TPS3 with FPP substrate.

Figure 23: GC-MS chromatogram of TPS3 with GPP substrate.

Figure 24: GC-MS chromatogram of TPS4 with FPP substrate.

Figure 25: GC-MS chromatogram of TPS4 with GPP substrate.

Examples

Example 1 - Isolation and characterization of tpsl and tps2 Genes

1. Distribution and terpene content of leaf glandular trichonies

The average amount of glandular trichomes per cm² of Origanum leaf surface was determined before creating cDNA libraries from isolated peltate glandular trichomes. Only the glandular trichomes at the secretory stages were counted due to the reflection of the subcuticular cavity. The secretory stage is the mature stage of gland development. Scanned (1200 dpi) leaves from different developmental stages showed a decrease in peltate glandular trichome concentration from younger to older leaves (Figure 4). The glandular trichomes are visible as small dark or white spots on the leaf surfaces.

Young, not completely extended leaves (< 1 week old) had two to seven times more glandular trichomes per cm² than old leaves (4 - 8 weeks). Differences were also detected between both lines of Origanum vulgare. The Origanum line f02-04 has 1.8-fold more glandular trichomes per cm² of young leaves than line d06-01. Similar results were found for pentane extractions from young leaves of both lines (Figure 5). Only half of the amount of total terpenes was extracted from line d06-01 in comparison to line f02-04 using the same amounts of leaf material. In both lines, γ- terpinene (peak no. 12) was the main terpene extracted.

The relative amounts of individual terpenes in young leaves of line fO2-O4 and d06-01 are compared in Table 2. Pentane extracts from older leaves showed only low amounts of terpenes per the same amount of dried leaf material (data not shown).

Monoterpenes were found in higher numbers and in higher concentration in both plant lines representing 73 % of all terpenes in line d06-01 and over 83 % in line f02-04. Altogether, 20 monoterpenes and 12 sesquiterpenes were extracted from both lines. Of these, only six terpenes showed significantly different proportions between both plant lines (Figure 6 and Table 2); cis- ocimene, allo-ocimene, ?ra»,s-β-caryophyllene and c/^'s-α-bisabolene were proportionally greater in line d06-01 whereas thymol and /rarø^nmϊ-α-farnesene were greater in line fi02-04.

γ-terpinene was the major terpene in both Origanum lines with 36 % in d06-01 and 32 % in f02-04. Terpenes which contributed three or more percent to the total terpene content added up to 84 % for d06-01 and 85 % for fϋ2-04, respectively.

Table 2: Individual terpenes (% of total) identified in pentane extracts from leaves of the d06-01 and f02-04 Origanum lines. Significant differences (p=0.05) between the two lines are highlighted in grey.

n.d.= not detected, camphene was found only in traces < 0.05 %.

Three of the terpenes were found in only one of the two Origanum lines; cώ-α-bisabolene was found only in Origanum line d06-01 and carvacrol and trans, tram-a-farnesene were found only in line f02-04. The biggest differences were detected for thymol which is a major monoterpene in line f02-04 (10.1 %) but only contributed 0.1 % in line d06-01. Sesquiterpenes contributed more to the total terpene content in Origanum line d06-01 due to the higher abundance of trans-Q- caryophyllene and l,6-germacradien-5-ol.

2. Isolation of terpene synthase-like genes from leaf gland cDNA libraries 2.1. Gland tissue preparation and library construction

The peltate glandular trichome preparation from young Origanum leaves was checked for contamination with other plant tissue fragments and hairy trichome cells from the leaf surface. Figure 7 shows isolated head cells of a peltate glandular trichome which had been abraded from the leaf surface. These head cells showed a clear partitioning into eight large peripheral cells and four small central cells. A cellular fraction enriched for these cells was used for RNA isolation with Trizol^® reagent to yield high concentrations of terpene synthase mRNA within the total RNA. A cDNA library in the plasmid pDNR-Lib and a GeneRacer™ library were created from the total leaf gland RNA from Origanum line f02-04. A SMART™ RACE cDNA library of Origanum line d06-01 was provided by Dr. Joerg Degenhardt.

2.2. RACE-PCR with the RACE cDNA library from Origanum line d06-01

The RACE 3' or 5' cDNA libraries from Origanum line d06-01 were used as template for RACE- PCR with the universal primer mix (UPM) and the gene specific oligonucleotide (GSP). The UPM binds to the SMART™ oligonucleotide which had been used for the reverse transcription of the mRNA. The first expressed sequence tags (ESTs) were obtained by this procedure with a GSP derived from the cDNA sequence of a gene encoding sabinene synthase from Salvia officinalis. These cDNA fragments were used to design new oligonucleotides primers which would have sequence specificity to Origanum vulgar e, Orelfwd and Orel rev. The two primers were used for most of the following RACE-PCR. All derived fragments were cloned into the pCR4-TOPO vector for sequencing. The sequences from the subcloned RACE-PCR fragments led to the isolation and sequence determination of two cDNAs which at first appeared to be one gene or alleles. These two genes were called tpsl and tps2. tpsl and tps2 had a similarity of 67.5 % at the nucleotide level and of 58.5 % at the level of the deduced amino acid sequence. Both genes had high similarities to terpene synthase genes for sabinene synthases and linalool synthases from different members of the Lamiaceae plant family as shown by the results from a BLAST search of the entire NCBI database via the internet (see table 5).

Those cloned genes were used to isolate further terpene synthase-like genes from Origanum libraries and in gene expression in both investigated Origanum lines.

2.3. Isolation of terpene synthase-like cDNA by sequencing of the pDNR-Lib primary cDNA library

A library of DNA sequences from individual clones of the primary cDNA library (ESTs) from Origanum line f02-04 was generated (Genomics Core Facility, Purdue University West Lafayette, Indiana, USA). The library had a titer of 5.4 x 10⁵ cfu/ml. The sequencing was performed only with the pDNRLfwd primer. Colony PCR with pDNRLfwd and pDNRLrev revealed a cDNA insert size distribution from 300 to 1,400 bp in this library with an average size of approximately 600 bp. BLAST searches in the NCBI database (http://www.ncbi.nlm.nih.gov/BLAST/) with the ESTs derived from sequencing of this primary library gave 42 hits for terpene synthase-like genes out of a total number of 2,304 ESTs. All those ESTs were assembled with each other as well as with other fragments from the RACE cDNA library and with tpsl and tps2 of the Origanum line d06-01. The analysis resulted in another 34 cDNA-fragments that had good matches (> 70 % identity) with terpene synthase-like genes. Most cDNA fragments assembled with the already identified tpsl or tps2 genes from the line d06-01. The remaining contigs (assembled ESTs) showed similarity to monoterpene or sesquiterpene synthases from other plants.

Table 3: Results of a BLAST search with assembled cDNA fragment contigs from the Origanum fD2-04 pDNR-Lib cDNA library.

3. Comparison of tpsl and tps2 sequences to related monoterpene synthases

The complete open reading frames (ORFs) of tpsl and tps2 from the Origanum line d06-01 as well as tps2 from the Origanum line f02-04 show (Table 4) very similar values for tpsl and tps2 in sequence length and predicted molecular weights of the protein.

Table 4: Properties of terpene synthase-like genes from Origanum. DNA and AA sequences are shown with GC content and predicted molecular weight, respectively.

The tpsl and tps2 sequences were compared with the terpene synthase sequences in the NCBI database and show (Table 5) between 55 - 60 % identity at the amino acid level with the four most closely related sequences found in the NCBI database.

Table 5: Sequence identity of tpsl and tps2 with related monoterpene synthases.

*Wise et al, 1998; **Crowe et a , 2002; **Yamada et a , irect submission to NCBI database. The identity between the two Origanum genes and the most closely related genes found in the NCBI database were not higher than 73.7 % on the nucleotide level. The sabinene synthase gene from Salvia officinalis revealed the highest similarities to almost all of the three terpene synthase- like genes from the Origanum vulgare cDNA libraries. Only tps2 from d06-01 showed a slightly higher similarity to the linalool synthase from Mentha citrata.

The protein sequences were also compared to those of the NCBI database. The similarities were much more obvious on the amino acid level. TPSl and TPS2 from both Origanum lines showed the highest similarity to the protein sequence of the sabinene synthase fvomSalvia officinalis with 59.2 % for TPSl and 62.3 % or 62.6 % for TPS2-d06-01 or TPS2-f02-04, respectively (Table 5).

All protein sequences showed common motifs for terpene synthase protein sequences including the double arginine, RR, at the N-terminal end of the protein and the aspartate-rich peptide, the DDXXD motif of the catalytic center in the C-terminal domain of the protein (Figure 8 and Figure 9).

The possibilities for signal peptide sequences in the polypeptides were predicted with the ChloroP, SiganlP and TargetP over the CBS Prediction Servers site in the internet (http://www.cbs.dtu.dk/services/). All three deduced protein sequences were predicted to have a signal peptide by two programs, but not by SignalP.

The algorithms predicted signal peptides of 24 AA for TPS2 from d06-01, 44 AA TPS2 from fO2- 04 and 35 AA for TPSl from d06-01. The longer signal peptide for TPS2 from line f02-04 is caused by the change at position 23 from valine (V) to methionine (M) (Figure 9).

The two TPS2 sequences showed an identity of 99.3 % on the nucleotide level and of 99 % in the amino acid sequences which means that only 13 DNA bases or 5 amino acids are different. All changes in the AA sequence are situated at the N-terminus within the first 262 amino acids.

4 . Heterologous expression of the tpsl and tps2 genes

The terpene synthase-like genes tpsl and tps2 were cloned into the pHis8-3 expression vector and heterologously expressed in E. coli BL21 (DE3).

The tpsl was expressed in 0.2 1 expression cultures. Desalted protein extracts from these small expression cultures were used in assays to test the enzyme activity and to determine the products formed by TPSl . Extracts from expression cultures with tpsl gene had enzyme activities that produced terpenes from the GPP substrate. Analysis of the assay extracts by GC-MS revealed sabinene as the main product, together with very small amounts of myrcene andγ-terpinene as well as traces of α-thujene and α-pinene (Figure 10).

The second terpene synthase-like gene, tps2, was similarly expressed and assays were performed in the same way as with TPSl . The results showed a high production ofγ-terpinene and nine other monoterpenes. The overall activity level of TPS2 was approximately 1.8-fold higher than that of TPSl. Also, TPS2 produced a greater variety of different monoterpenes. Ten monoterpenes were found altogether: α-thujene, α-pinene, sabinene, β-pinene (4a), myrcene, α-phellandrene, α- terpinene, p-cymene, limonene (8a) and γ-terpinene (Figure 11).

Table 6: Commercially available terpene standards and their origin.

TPSl was named a sabinene synthase and TPS2 a γ-terpinene synthase. Both TPSl and TPS2 will be used synonymously with their trivial names in further descriptions.

5. Biochemical characterization of the monoterpene synthase TPS2 5.1 Heterologous expression and purification of the TPS2 protein The enzyme activity of TPS2 was characterized in more detail, because aγ-terpinene synthase has not been isolated from Oregano before. TPS2 was expressed as a full-length protein in large scale bacterial cultures to obtain enough purified protein for a biochemical characterization. TPS2 was purified from the bacterial proteins of the E. coli strains. It was expressed in utilizing the N- terminal His-tag which was introduced with the pHis8-3 expression vector. Therefore, the purified protein was 18 AA longer than the native protein. This AA fragment contains a cleavage site for thrombin to cleave off the His-tag. Cleavage of the His-tag was performed with the first preparations of purified protein. The additional purification step was necessary to divide cleaved and uncleaved protein as well as to remove the cleaved sequence tag but resulted in high losses of protein. This residual amount of protein has not been sufficient to perform all assays for the biochemical characterization. Therefore, all following assays for biochemical characterization were performed with the His-tagged protein. Protein purification was performed from 4 1 of bacterial cultures in TB media which gave approximately 10 g of bacterial cells per liter of media.

A concentration of approximately 350 μg protein per ml was measured after purification with Ni- NTA agarose (Qiagen, Hilden, Germany), dialysis to remove imidazole and reduction of the NaCl concentration to 100 mM. The total yield of protein was 3.15 mg.

The predicted size of the purified protein was 70.9 kDa including the His-tag. This corresponds to the size determination with an SDS-gel which revealed a protein band in the range of 70 kDa.

5.2 Characterization of the TPS2 enzyme activity

The main product of TPS2 was γ-terpinene with 79.9 % of the total terpenes; two other terpenes contributed more than five percent (α-thujene 7.2 % and α-terpinene 6 %) and the last 6.9 % were divided among seven terpenes (myrcene 2.6 %, sabinene 1.3 %, limonene 1.2 %, α-pinene 1 % and α-phellandrene 0.75 % and traces of β-pinene and /?-cymene). TPS2 produced only traces of sesquiterpenes with FPP substrate and enzyme assays with GGPP (GPP, Geranyl pyrophosphate, GGPP=Geranylgeranyl pyrophosphate) substrate revealed no terpene products.

First, the relative enzyme activity of the expressed protein over time was investigated. The cumulative product formation showed a higher increase during the first 60 minutes (Figure 12) than within the second hour. In the first 60 minute interval, the terpene amount increased by 1.0 to 4.5 ng terpenes per minute and μg protein with its highest increase in the 0 to 10 minutes interval. Further increase was in a range between 0.01 to 0.6 ng terpenes per minute and μg protein. At the 60 minute time point, 85 % of the total accumulated terpenes (determined after 120 minutes) were formed.

For all following experiments, an assay time of 60 minutes was used to gain enough terpenes in the extracts for GC-FID detection.

Two different protein concentrations were used to determine K_m values for the GPP (geranyl pyrophosphate) substrate. Commonly, terpene synthases use Mg²⁺ and Mn²⁺ as cofactors. Therefore, the activity in the presence of these two metal ions and their K_m values was determined. Furthermore, the activity levels for different temperatures and at different pH values were measured.

The K_n, values were calculated by the EnzymeKinetics module of the SigmaPlot program. These values have been compared with values derived from regression of double reciprocal graphs after the method of H. Lineweaver and D. Burk (1934). Values derived from both methods are shown in Table 3.5. The main terpene product γ-terpinene was used for all calculations. All values have been standardized with a known amount of the internal standard (IS), the sesquiterpene δ-cadinene.

The K_m values for the GPP substrate are in a similar range for both methods. The values for the metal ions show greater differences, especially for Mg .2+

Table 7: K_m values for GPP, Mg²⁺ and Mn²⁺ and their standard error.

1.75μg protein per assay was used for all but the GPP (2) set of assays with 3.5 μg protein per assay. Mn²⁺ was the metal ion in all assays excluding the Mg²⁺ assays. V_max values are given for 1 μg of protein. The two different protein concentrations (1.75 and 3.5 μg) show similar kinetics (Figure 13). A high increase could be seen for substrate concentrations in the range from 0 to 10 μM, followed by a reduced increase of the slope up to 30 μM GPP where the asymptotic area was approached.

The enzyme activity in the presence of Mn²⁺ ions shows a very strong increase at low concentrations of manganese ions up to 0.5 μM where the slope reaches a plateau (Figure 14). At higher manganese ion concentrations, the enzyme activity decreases continuously caused by an inhibition of enzymatic activity at concentrations above 1 μM.

The enzyme activity rises more slowly with increasing concentrations of magnesium ions (Figure 15). It reaches the plateau at a concentration of about 20 μM Mg²⁺ and decreases continuously until no more enzyme activity could be found at a concentration of 200 μM Mg²⁺.

A comparison between both metal ions shows an approximately 2-fold higher enzyme activity in the presence of manganese (V_max). The highest activity with manganese was measured at concentrations of 0,5 μM instead of 10 μM for magnesium which is a 20-fold higher concentration.

The determination of enzyme activity at different pH values shows an optimum between 6.0 and 8.5 (Figure 16). The highest enzyme activity was found for a pH of 6.8. Half maximal enzyme activity was found at pH 7.2 to 7.5 and pH 6.0 to 6.5.

The enzyme activity at different temperatures shows a maximum at 28 ⁰C and half maximal activity at 20 and 35 ⁰C (Figure 17). Enzymatic activity was also detected at 4 ⁰C and 42 ⁰C under in vitro conditions.

The enzyme could be stored on ice for a period of 4 days without significant loss of enzyme activity. Freezing at -20 and -80 ⁰C reduced the enzymatic activity drastically to less than the half.

5.3 Comparison of original plant terpene composition with the blend of TPS2 in vitro products

The proportions of individual terpenes produced by TPS2 were compared to those generated from pentane extractions of the original plant leaves which were considered as 100 % for the same group of terpenes. Both the heterologously expressed and purified enzyme and the plant pentane extracts revealed a very similar composition of terpenes (Figure 18).

Only sabinene was found in significantly higher (p=0.05) concentrations in plant pentane extracts than in the TPS2 in vitro enzyme assay products. Conversely, the γ-terpinene synthase produced β- pinene (4a) and limonene (8a) in vitro which could not be found in plant pentane extracts. The purified enzyme produced α-thujene and γ-terpinene in significantly higher (p=0.05) amounts than they were found in the plant pentane extracts.

5.4 Transcript levels of tps2 in both investigated Plant lines

The expression of tps2 was measured in both lines to test whether the absolute levels ofγ-terpinene production correspond to differences in tps2 transcript levels.

RNA blots containing five μg of RNA isolated from 100 mg young leaves of both lines were hybridized with a radiolabeled 1 kb fragment from the 3 ' end of tps2 from line dO6-01. No significantly different tps2 transcript levels were found in young leaves of both lines (Figure 19).

6. Discussion

6.1 The composition of terpenes in original plants

Thyme, oregano, sage, basil and others, economically important herbs, contain essential oils which consist mostly of mono- and sesquiterpenes. In Oregano, terpene production is located in glandular trichomes on the leaf surface. The composition of the terpene components in the essential oil and in its biosynthesis can therefore be studied in these glandular trichomes.

The density of peltate glandular trichomes was high on emerging leaves and decreased during their development.

Two clonal plant lines, d06-01 and f02-04, with different terpene content and composition were investigated. The terpene composition of the two lines differed only for six out of 32 terpenes, cw- ocimene, allo-ocimene, thymol, fraws-β-caryophyllene, trans, trans- α-farnesene and cis-a- bisabolene. The largest variation was found for thymol which provided either 0.13 % (d06-01) or 10 % (f02-04) of the total terpene content. Compared to other measurements of essential oil composition in other herb species, these two ines contained relatively low amounts of thymol and carvacrol. Values of over 50 % of the total terpene content are often reported for these terpenes. Essential oil measurements of the line d06-01 revealed a large variation of the />-cymene concentration between independent harvests which is reflected in a high standard error.

6.2 Identification of terpene biosynthesis from cDNA libraries

To more easily isolate the cDNAs corresponding to the terpene synthases, RNA was extracted from tissue types where the terpenes are synthesized, namely the peltate glandular trichomes. The screening of cDNA libraries is a convenient method to identify genes of terpene biosynthesis from mRNA. Therefore several cDNA libraries were constructed out of RNA isolated from peltate glandular trichomes.

First, cDNA fragments were obtained from a RACE cDNA library of line d06-01 with PCR using oligonucleotide primers with degenerate sequences corresponding to the cDNA sequence encoding a sabinene synthase from a sage herb. Two monoterpene synthase-like genes, tpsl and tps2, were identified and facilitated the identification of further terpene synthase gene candidates from a second cDNA library of line f02-04. The large-scale sequencing of this second cDNA library revealed 76 expressed sequence tags (ESTs) out of 2,304 ESTs which could be related to mono- and sesquiterpene synthase genes and one cytochrome P450 monoxygenase gene.

These ESTs can be further divided into 41 ESTs (54 %) with a high sequence identity to previously identified monoterpene synthase genes (> 80 %), 30 ESTs (39.5 %) closely related to sesquiterpene synthase genes (identity > 80 %) and the last five ESTs (6.5 %) showed a sequence identity of 81 % to a cytochrome P450 limonene-hydroxylase from Mentha sp. which was reported to convert 4S- limonene into (-)-£ra«.y-isopiperitenol or (-)-/ra«5-carveol in peppermint and spearmint, respectively (Lupien et al., 1999). Cytochrome P450 monoxygenases are reported to play an important role in secondary transformations of terpenes (Gershenzon and Croteau, 1993, Wise and Croteau, 1999).

Assembly of the ESTs from identical genes into larger cDNA fragments showed that two cDNAs were represented by a high number of ESTs. This suggests that the two corresponding mRNAs were present in high concentrations in the glandular trichomes. Nineteen ESTs assembled to form a 1,320 bp contiguous fragment with similarity to a δ-cadinene sesquiterpene synthase from cotton. Since only low concentrations of sesquiterpenes are present in the essential oil of herbs, this gene was not further investigated because the beginning of the open reading frame was missing. However, the longest fragment of 14 ESTs with 2,351 bp revealed a 99 % amino acid identity to tps2 from line d06-01. It might therefore be an allele of tps2 or a very closely related member of the monoterpene synthase gene family.

The total number of 76 ESTs and 11 discrete alignments indicate a final number of four to five different monoterpene synthases, up to four sesquiterpene synthases and one cytochrome P450 monoxygenase in line f02-04 that are transcribed and might play a role in essential oil synthesis in glandular trichomes.

The sequencing results from the primary cDNA library of line f02-04 are very promising for the discovery of further mono- and sesquiterpene synthases and cytochrome P450 monoxygenases. Further investigations will be needed to understand the role of these ten enzymes in the biosynthesis of the 32 terpenes in these two lines.

6.3 Sequence characteristics and enzymatic profiles of TPSl and TPS2

The sequences of the complete open reading frames (ORFs) oitpsl and tps2 from line d06-01 and of tps2 from line f02-04 were found to be closely related with monoterpene synthases of the Lamiaceae family. The comparison of the deduced protein sequences of these genes and 46 previously identified monoterpene synthases reveals a clear division into several subgroups of terpene synthases and shows a close relation of monoterpene synthases in the Lamiaceae (Figure 20).

A classification by Bohlmann e? al. (1997) divided plant terpene synthases into the six subgroups Tpsa to Tpsf which were initially defined by a minimum of 40 % amino acid identity between group members. Monoterpene synthases from the Lamiaceae were predicted to belong to the Tpsb subgroup and both the sabinene synthase (tpsl) and the γ-terpinene synthase (tps2) identified in this study clearly belong to this subgroup. A putative γ-terpinene synthase from line d27 has the closest relation to the γ-terpinene synthase from d06-01 (unpublished).

The second largest group of genes in the dendrogram is the group of monoterpene synthases from gymnosperms (Tpsd) which form a homogenous subgroup distant to terpene synthases from angiosperms. This large distance was proposed to be caused by an early separation of terpene synthases of angiosperms and gymnosperms from a common ancestor (Bohlmann et al., 1998).

The terpene synthases from Citrus limon and Citrus unshiu also form a subgroup which contains three γ-terpinene synthases which form products similar to TPS2. They form γ-terpinene and four of the same minor terpene products of TPS2 but in different proportions (Table 8). Although these three γ-terpinene synthases form the same major products as TPS2, they display a very low amino acid sequence identity of 37.1 to 37.6 %.

Table 8: Terpene product profiles of TPSl-d06-01 and TPS2 and monoterpene synthases forming similar products.

The sabinene synthase, TPSl-d06-01, had a 59 % amino acid identity with a previously identified sabinene synthase from Salvia officinalis (Wise et al, 1998). Comparison of TPSl-d06-01 enzymatic products with those of the sabinene synthase from S, officinalis showed a higher relative sabinene production (85.2 %) by the Origanum vulgare enzyme. Also, TPSl produced a higher number of minor terpene products than the sabinene synthase from S. officinalis which was reported to produce only γ-terpinene as an additional terpene.

6.4 Biochemical characterization of TPS2

Multiproduct formation by terpene synthases was initially thought to be caused by incompletely purified enzymes from plant extractions. Further purification of terpene synthase activities and especially heterologous expression in bacterial expression systems led to the observation that these enzymes can form multiple terpene products even when pure (Wise and Croteau, 1999).

TPS2 was identified as a multiproduct enzyme that forms γ-terpinene and nine other related monoterpene products. The amino acid sequence of TP S2 showed a high similarity to other monoterpene synthases especially those of other Lamiaceae. The K_m values for the GPP substrate (8.7 μM) and the metal ions Mn²⁺ and Mg²⁺ (0.07 μM and 3.41 μM) conform with those of other angiosperm monoterpene synthases which were determined to be 1.1 - 7.7 μM for GPP and 0.3 - 5 μM for the preferred metal ion (Mn²⁺ or Mg²⁺) (Wise et al, 1998, Williams et al, 1998, Croteau et al, 1994). The K_m values for the metal ions indicate a clear preference for manganese which might be an important cofactor in planta.

The pH optimum coincides with the pH optima of other monoterpene synthases which are in the range of 6.0 to 7.2. The limonene synthase from peppermint was reported to be located in the stroma of plastids of the secretory cells (Turner et al, 1999). Since the pH in the stroma of spinach chloroplasts is reported to be 7.4 - 7.9 (Oja et al, 1999, Wu and Berkowitz, 1992) the terpene synthases might not be in an environment with optimal pH in vivo. However, the enzymatic activity of TPS2 is still at half of the in vitro maximum at pH 7.5 and can be active in terpene production. It also is possible that the pH value in Origanum plastids is lower than those of spinach.

The temperature the plant is exposed to is another important factor regulating enzyme activity. The optimal temperature for TPS2 activity was determined to be 28 ⁰C which might be due to the fact that the main distribution of the herb species is in the Mediterranean region. The very broad temperature range of the TPS2 enzyme activity from 4 ⁰C to 42 ⁰C might allow the biosynthesis of essential oil components under a wide range of climatic conditions. A limited number of temperature optima for other monoterpene synthases have been reported in the literature.

One aspect of the biochemical characterization of the γ-terpinene synthase was surprising: TPS2 also produces minor amounts of limonene and β-pinene which were not found in leaf extracts although these compounds have been reported for other lamiaceae herb essential oils in minor amounts (Russo et al, 1998, D'Antuono et al., 2000). Limonene contributed 1.2 % to the total terpene products of TPS2 and β-pinene 0.5 %. The occurrence of these compounds could be an artifact of the in vitro assay conditions for TPS2 but it is also possible that these products are formed and metabolized further by additional enzymes in planta. The conversion of limonene to other metabolites was described for a cytochrome P450 limonene-6-hydroxylase from peppermint and spearmint (Lupien et al, 1999). The possibility of a hydroxylation of limonene in lamiaceae herbs was supported by the presence of cDNA in the cDNA library from line f02-04 which had a high sequence identity to the limonene-6-hydroxylase of spearmint and peppermint (Table 4.1). But neither of the two hydroxylation products, fr-β/w-isopiperitenol and trans-carveo\ which were reported for the limonene-6-hydroxylase of Mentha sp. were found in lamiaceae herb essential oils. 6.5 The predicted reaction mechanism of TPS2

Extensive studies on the monoterpene synthases of Mentha species showed that all monoterpenes are derived from the Geranylpyrophosphate (GPP) substrate via a carbocationic reaction mechanism (Wise and Croteau, 1999). Figure 21 shows the predicted reaction intermediates in the biosynthesis of TPSl and TPS2 terpene products. The scheme is based on the reaction mechanisms described by Wise and Croteau (1999).

The acyclic GPP substrate is dephosphorylated and forms a carbocation, the linalyl cation (see Introduction). This early intermediate is cyclized to form the α-terpinyl cation which is the most central intermediate in the reaction mechanism. The acyclic product myrcene is derived from the linalyl cation by deprotonation before the cyclization of the α-terpinyl cation. Limonene is then formed by deprotonation of the α-terpinyl cation intermediate. The main terpene, γ-terpinene, and six minor products are derived by hydride shifts which lead to the formation of further intermediates. The terpinen-3-yl cation is the precursor forα-thujene and the terpinene-4-yl cation is the precursor for α-terpinene and γ-terpinene. The biosynthesis of sabinene requires the formation of the sabinyl cation intermediate derived from the terpinene-4-yl cation. Bothα-pinene and β-pinene are derived from a third major intermediate, the pinyl cation. The aromatic monoterpene j?-cymene is catalyzed in an additional step by desaturation of γ-terpinene. /?-Cymene is considered the precursor for thymol which is derived from hydroxylation of/7-cymene (Poulose and Croteau, 1978). Thymol plays an important role in herb essential oil but it was found only in minor amounts in line d06-01, the source of TPS2.

6.6 Regulation of terpene production and TPS2 transcript levels in the two Origanum lines

The terpene concentration in leaf extracts of line d06-01 was only half of that of line f02-04. This difference in terpene concentrations corresponds to the 50 % lower density of glandular trichomes on leaves of line d06-01. Thus, it was expected to find lower levels of γ-terpinene synthase mRNA in line d06-01. The RNA was extracted from whole leaves. However, the TP S2 transcript levels in line d06-01 and line f02-04 revealed identical transcript levels. This indicates that the TPS2 transcript concentration does not influence the level of terpene formation in the two lines. These data do not indicate whether TPS2 is an important enzyme for terpene formation in planta. If TPS2 is indeed involved in in vivo essential oil biosynthesis, it could be speculated that the terpene biosynthesis is regulated at a later step such as protein translation. Nevertheless, cross hybridizations between the TPS2 probe and mRNA of other terpene synthases in the Northern blot assay cannot be excluded. It will therefore be important to verify the Northern blot results by quantitative PCR methods.

6.7 TPS2 and its function in essential oil biosynthesis

The multiple terpene products of TPS2 correspond very clearly to the terpene blend extracted from line d06-01 which was used to isolate this terpene synthase. Eight of the ten TPS2 products were found in the plant extracts where they represent 50 % of the total terpene content. The relative proportions of these terpenes are similar in the TPS2 product blend and the plant's essential oil. This shows that TPS2 is responsible for a large part of essential oil biosynthesis in line d06-01.

The second monoterpene synthase, TPSl, produces mainly sabinene which is another important compound of the essential oil and contributes 7 % to the total terpene content. Thus, 57 % of the terpenes can be attributed to these two enzymes.

6.8 Generation of transgenic Arabidopsis plants expressing tps2

The open reading frame of tps2 will be amplified from the bacterial expression vector by PCR and inserted as a 5αmHI-KpnI fragment between the 35 S promoter of the Cauliflower mosaic virus (CaMV) and a nopaline synthase terminator into the binary vector pBIN420 (Browse et. al., 2001). The obtained constructs are introduced into the Agrobacterium tumefaciens GV3101 strain, which is used to transform Arabidopsis (ecotype Col) plants by the floral dip method (Clough et. al,, 1998). Transgenic lines are selected on kanamycin resistance and the presence of tps2 insertion into the resistant plants is additionally confirmed by PCR analysis. The transgenic plants will be grown under short day conditions (8 h light) for four weeks until they reach the rosette stage. The transformants will be screened for emission of tps2 monoterpene products. We expect to find Arabidopsis transformants with a high constitutive tps2 product formation at the rosette stage. Control lines will be established carrying the empty insertion cassette only.

6.9 DNA and amino acid sequences

This paragraph contains the complete open reading frames and their deduced amino acid sequences of all heterologously expressed monoterpene synthase genes from Origanum vulgar e. 6.9.1 DNA sequences

Seq. 7.1 : ORF DNA sequence of TPSl-d06-01, sabinene synthase (SEQ ID NO: 5)

Seq. 7.2: ORF DNA sequence of TPS2-d06-01, γ-terpinene synthase (SEQ ID NO: 1)

Seq. 7.3: ORF DNA sequence of TPS2-fϋ2-04, γ-terpinene synthase (SEQ ID NO: 2)

6.9.2 Amino acid sequences

Seq. 7.4: Deduced AA sequence of TPSl-d06-01, sabinene synthase (SEQ ID NO: 6)

Deq. 7.5: Deduced AA sequence of TPS2-d06-01, γ-terpinene synthase (SEQ ID NO: 3)

Deq 7.6: Deduced AA sequence of TPS2-f02-04, putative γ-terpinene synthase (SEQ ID NO: 4)

Example 2

A Gamma-Terpinene Synthase was isolated from Thymus vulgaris, Chemotype T28, as was described above for TPS2. The Enzyme is structurally related to TPS2 and when expressed in E. coli, leads to the production of the same terpenes like TPS2, under identical experimental conditions.

gamma-terpinene-proteine (T28): (SEQ ID NO: 7)

gamma-terpinene-sequence (T28): (SEQ ID NO: 8)

Example 3

Analysis of two sesquiterpene synthases from Origanum vulgare L.

Two sequences from putative sesquiterpene synthases of Origanum vulgare L. have been cloned into the pHis8-3 expression vector for heterologous expression in Escherichia coli BL21 (DE3) (Invitrogen, USA).

Enzyme assays were performed with crude extracts from heterologously expressed sesquiterpene synthases of Origanum vulgare L. and analyzed by GC-MS. 1. Protein expression and enzyme activity analysis

1.1. Heterologous expression of terpene synthase genes in E. coli

Positive colonies were picked to set up 5 ml overnight cultures in LB medium with appropriate antibiotic. The next day, the grown cells were checked for the presence of the insert with a variation of colony-PCR using 2 μl of overnight cultures as template and gene specific primer pairs.

These 5 ml precultures were used to inoculate 500 ml Erlenmeyer-flasks with 200 ml TB containing 50 μg/ml kanamycin. Cultures were grown to an ODβoo of 0.6 to 0.7 at 37 ⁰C and 220 rpm. Transcription was induced at this point by addition of 0.8 M IPTG solution (0.5 mM final concentration). After induction cultures were grown for 18 h shaking at 18 ⁰C and 220 rpm.

1.2. Extraction of the expressed protein from 0.2 1 cultures

Cultures were harvested by spinning the bacteria down at 6,000 x g for 5 min in an AVANTI 25 centrifuge (Beckmann, Fullerton, US). The supernatant was removed and the remaining cell pellet was resuspended in 3 ml of extraction-buffer (50 mM MOPSO (pH 7), 10 % glycerol, 5 mM DTT, 5 mM Na-ascorbate, 0,5 mM PMSF). The cells were disrupted by sonication with the sonicator Sonoplus UW 2070, 4 min at 60 % (Bandelin electronics, Berlin, Germany).

The sample was centrifuged at 16100 xg for 45 min in a Centrifuge 5415 R (Eppendorf, Hamburg, Germany) to separate protein and cell debris.

The supernatant was collected and transferred to assay buffer (20 mM MOPSO (pH 7), 10 % glycerol, 1 mM DTT) using Econo-Pac 10DG columns (BioRad, Hercules, USA) which had been previously equilibrated with 10 ml of assay buffer. According to the manufacturer's manual 3 ml of the supernatant was applied on top of the column, allowed to run through and eluted with 4 ml of assay buffer. The 4 ml flow-through was collected for further use in protein assays. For short storage, the protein extract was frozen at -20 ⁰C and for longer storage at -80 ⁰C.

1.3. Assays for terpene synthase activity

The activity of purified terpene synthase was measured in assays with GPP and FPP substrate. The enzyme assays were performed in 2.5 ml glass vials for GC analysis. 1 ml standard assay with the final concentration indicated in parentheses:

765 μl Assay puffer (20 niM MOPSO, pH 7, ImM DTT, 10% v/v Glycerol)

10 μl FPP/GPP (5 μM)

5 μl MnCl₂ (1 μM)

20 μl MgCl₂ (10 μM)

200 μl crude extract from E. coli cultures

Assays were overlaid with 200 μl n-pentane and then incubated for 60 min at 30 ⁰C in a GFL 1002 water bath (Gesellschaft fur Labortechnik, Burgwedel, Germany). The glass vials were shaken on a small incubation shaker (Eppendorf, Hamburg, Germany) 2 times for 1 min at 1,400 rpm to stop the assay and to partition the terpene volatiles to the solvent phase. To remove all water from the solvent phase, the glass vials were deep frozen for 45 min at -80 ⁰C. 130 μl of solvent phase were then transferred to a new 2 ml glass vial with glass insert for further analysis by d GC-MS.

GC-MS analysis was performed on a HP 6890 GC System with MS detector (HP 6980 GC System with 5973 Network Mass Selective Detector). A DB5-MS column (30 m length, 0.25 mm inner diameter and 0.25 μm film by J&W Scientific, Folsom, USA) was installed. For injection of samples, an HP 7683 Series Injector was used. Helium (1 ml/min) was used as carrier gas for GC- MS. Injection temperature was 230 ⁰C.

1.4. Analysis of enzyme activity assays

Analysis of terpenes was performed with the following conditions: Sample volume: 1 μl (splitless mode) Program for the oven heating:

3 min 40 ⁰C

Ramp l 6 °C/min 40 - 185 ⁰C

Ramp 280 °C/min 185 - 250 ⁰C

3 min 240 ⁰C

2. Results and Discussion

2.1. Sequences and enzyme activity of sesquiterpene synthases

2.1.1. DNA sequences

Sequence 2.1.: Open reading frame DNA sequence of TPS3 (SEQ ID NO: 9)

Sequence 2.2.: Open reading frame DNA sequence of TPS4 (SEQ ID NO: 10) 2.1.2. Protein sequences

Sequence 2.3.: Putative protein sequence of TPS3 ORF (SEQ ID NO: 11)

Sequence 2.4.: Putative protein sequence of TPS4 ORF (SEQ ID NO: 12)

2.2. Enzyme activity and products of TPS3 and TPS4

Enzyme assays with the crude extract from E. coli expression cultures showed activity for both expressed enzymes for FPP and GPP substrate.

The sequence information together with comparison to other sequences showed higher similarities to other sesquiterpene synthases especially from the mint plant familiy. Highest similarities were found for a germacrene D synthase from Ocimum basilicum with 70.4 % identity in the amino acid sequence for TPS3 and 48.6 % identity for TPS4 and DNA sequences showed 76.2 % identity for TPS3 and 59.5 % identity for TPS4.

TPS3 and TPS4 had 62.8 % identity in the DNA sequence and 51.2 % at the amino acid level to each other. Both sequences do not have a N-terminal plastid targeting signal and are therefore cytoplasmatic sesquiterpene synthases.

The major product of TPS3 is germacrene D for FPP substrate and linalool for GPP substrate. TPS4 was supposed to be inactive since in the conserved DDxxD domain the third D was missing (Sequence 2.4.). Nevertheless, activity could be measured for FPP and GPP substrate for TPS4.

Claims

P A T E N T C L A I M S

1. A terpene synthase, which is encoded by the nucleic acid of SEQ ID NO: 1 or 2 or variants thereof, which variants are each defined as having one or more substitutions, insertions, and/or deletions as compared to the nucleic acid of SEQ ID NO: 1 or 2, provided that:

a) these variants hybridize under moderately stringent conditions to a nucleic acid, which comprises the sequence of SEQ ID NO: 1 or 2, and further provided that these variants code for a protein having terpene synthase activity; or

b) these variants have nucleic acid changes which are due to the degeneration of the genetic code, which code for the same or functional equivalent amino acid as the nucleic acid of SEQ ID NO: 1 or 2.

2. An isolated nucleic acid, which comprises the nucleic acid of SEQ ID NO: 1 or 2 or variants thereof, wherein the variants are each defined as having one or more substitutions, insertions, and/or deletions as compared to the nucleic acid of SEQ ID NO: 1 or 2, provided that:

a) these variants hybridize under moderately stringent conditions to a nucleic acid, which comprises the sequence of SEQ ID NO: 1 or 2, and further provided that these variants code for a protein having terpene synthase activity; or

b) said variants have nucleic acid changes which are due to the degeneration of the genetic code, which code for the same or functional equivalent amino acids as the nucleic acid of SEQ ID NO: 1 or 2.

3. The isolated nucleic acid of claim 2, which is further operably linked to one or more regulatory sequences.

4. A nucleic acid, which is a transcriptional product of one of the nucleic acids of claims 2 or 3.

5. A nucleic acid, which selectively hybridizes to transcriptional products of claim 4 under moderately stringent conditions.

6. A primer for the amplification of the nucleic acid of claim 2 or of a transcriptional product thereof.

7. A vector, which comprises the nucleic acid of claim 2 or 3.

8. An expression vector, which comprises the nucleic acid sequence of claim 2 or 3 and one or more regulatory sequences.

9. The vector of claim 8 which is a plasmid.

10. The vector of claim 9, which is the tumor inducing (Ti) plasmid.

11. A host cell, which has been transformed with the vector of claim 9 or 10 or which is containing the nucleic acid of claim 2 or 3.

12. The host cell of claim 10, which is a procaryotic cell.

13. The host cell of claim 12, which is E.coli or B. subtilis.

13. The host cell of claim 11, which is an agrobacterium.

14. The host cell of claim 13, which is Agrobacterium tumefaciens.

15. A terpene synthase, comprising the amino acid sequence of SEQ ID NO: 3 or 4 or a variant of said amino acid sequences, which variant comprises one or more substitutions, insertions, and/or deletions as compared to the sequence of SEQ ID NO: 3 or 4, and wherein the biological activity of the variant is substantially equal to the activity of the terpene synthase comprising the unmodified amino acid sequence of SEQ ID NO: 3 or 4.

16. An inhibitor of the terpense synthase of one or more of the preceding claims, wherein said inhibitor is selected from the group consisting of a molecule that reduces the level of mRNA encoding said terpene synthase, a molecule that reduces the level of said terpene synthase protein, and a molecule that reduces the biological activity of said terpene synthase.

17. The inhibitor of claim 16, wherein said inhibitor is selected from the group consisting of an antisense nucleic acid, a ribozyme, double stranded RNA, preferably siRNA, an antibody, a peptide and a peptidomimetic.

18. A method of producing a transgenic plant comprising the steps of: a) providing a nucleic acid of claim 2 or 3, or a vector of one of claims 7-10; b) transforming plant cells or tissues with said nucleic acid or vector; and c) generating whole plants from said transformed plant cells or tissues.

19. The method of claim 18, wherein the vector in a) is a Ti plasmid, wherein the tumor inducing sequences were replaced by the nucleic acid of claim 2 or 3.

20. The method of claim 19, wherein the plant cells or tissues are transformed by infection with A, tumefaciens containing the Ti plasmid.

21. The method of claim 18, wherein the transformation in step b) is performed by means of the gene canon method.

22. The method of one or more of claims 18-21, wherein the plant cell or tissue is derived from terpene producing plants.

23. The method of claim 22, wherein the plant is selected from sage (Salvia officinalis), oregano (Origanum vulgare), thyme (Thymus vulgaris) or basil (Ocimum basilicum).

24. A transgenic plant obtainable by the method of one or more of claims 18-23.

25. A method of producing one or more secondary plant metabolites comprising the steps of: a) providing a transgenic plant as defined in claim 24, b) culturing the plant under suitable conditions; and c) recovering said one or more secondary plant metabolites from said plant.

26. The method of claim 25, wherein the secondary metabolite is selected from the group of terpenoids.

27. The method of claim 26, wherein the terpenoids are comprising hemi-terpenes, mono- terpenes, sesquiterpenes, di-terpenes, tri-terpenes, tetra-terpenes and poly-terpenes.

28. An essential oil, which is obtainable by the method of one or more of claims 25-27.

29. A pharmaceutical composition comprising a therapeutically effective dose of the essential oil of claim 28 in combination with a pharmaceutically acceptable carrier.

30. A food or cosmetic product comprising the essential oil of claim 28.